The founder of a popular new website designed to help physicians understand a patient's prognosis says the information is not really directed to oncologists. But some oncologists may find it to be quite useful.
The site, ePrognosis (eprognosis.com) compiles several geriatric prognostic indices to help inform physicians about possible mortality outcomes. The indices are for older adults who do not have a dominant illness that is likely to be the cause of death, says Sei J. Lee, MD, a geriatrician at the University of California, San Francisco.
Because cancer is typically seen as a dominant disease—and often is a terminal illness—the mortality indices on ePrognosis “may be less relevant for oncologists,” he said.
Steven Tucker, MD, an oncologist in Singapore, begs to differ. He said that as a prostate cancer specialist, he was excited to see that ePrognosis includes the Lee Four-Year Index for Community-Dwelling Adults, which he has used for years to help reduce his patients' fear of dying from prostate cancer.
That index shows that a 65-year-old man with no known health problems has a six to nine percent risk of death in the next four years. (Lee SJ et al: JAMA 2006;295:801-808) By contrast, Tucker points out, a man newly diagnosed with prostate cancer, regardless of the stage, has a risk of death of less than one percent in the next five years.
“We know how frightening the word ‘cancer’ is,” he said. “But when I see a patient and think ‘You're overweight, you're smoking, you're diabetic, and you have a stent,’ I want to say, ‘Prostate cancer is not the problem.’”
Prognostic information that applies to conditions other than cancer can help oncologists consider what he calls “competitive mortality.” In many patients, cancer is competing with other factors—chronic disease, old age, and side effects from overly aggressive treatments—that all influence a patient's risk of death.
“Why would I chase the possibility of low-risk prostate cancer in someone who has triple-vessel coronary disease or poorly controlled diabetes?,” he said. “Let's stick to what we can fix first.”
Lee, the ePrognosis founder, developed the Lee Four-Year Index, but it is only one of 16 mortality indices included on the website. The indices were each developed for a specific population; for example, one index shows one-year mortality prognosis for newly admitted nursing home residents age 65 or older, one is for persons 65 and older who have resided in a nursing home for at least one year, and one is for hospitalized adults who are age 70 or older.
“It's your one-stop-shop if you are interested in mortality indices,” he said.
He and his co-developers—Alex Smith, MD, and Eric Widera, MD—are all geriatricians at UCSF. They, along with some colleagues, published a review of the 16 indices in JAMA earlier this year (Yourman et al: JAMA 2012;307:182-192).
The team's goal is to encourage physicians to consider a patient's prognosis as a key component of medical decision-making—and thus improve their patients' care.
The timeliness of referrals to hospice care, for example, might improve if physicians and patients had frank discussions about prognosis.
“Nearly everyone agrees that for hospice to optimally benefit patients, they should be enrolled in hospice for a timeframe of several months rather than several weeks,” Lee said. “The reason referrals to hospice generally happen so late is because patients and providers are uncomfortable thinking about life expectancy and prognosis.”
He and his colleagues also think that many healthy older adults do not get appropriate cancer screenings—for example, he said, patients with advanced dementia and other progressive diseases are sometimes screened for, diagnosed, and treated for cancers that would probably cause no symptoms before the patient died from another condition.
The main point is summed up in the introduction of the JAMA article, where the authors state: “Failure to consider prognosis in the context of clinical decision making can lead to poor care.”
That is exactly why Tucker, the prostate cancer specialist, keeps the mortality graphs and the patient questionnaire for the Lee index on the computer in his examining room.
“I bring it up all the time to show people the graphs,” he said. “I worry about my patients' health comprehensively, and it is tools like this that help me convince people that they have to make lifestyle changes, that they have to embrace exercise, better nutrition, meditation, better sleep, they need to listen to their cardiologist, take their aspirin, etc. And then we still deal with prostate cancer or elevated PSA, but we put it in perspective.”
Do Patients Care?
While physicians may be reluctant to focus on prognosis, people without a medical degree love to do so. In the first week after the ePrognosis website launched, it received more than 500,000 page views from 150,000 unique visitors.
Dr. Lee said he was not surprised—-because he experienced a barrage of attention after he published the Lee index six years ago.
“I was stunned at that point by the level of media interest on mortality prediction,” he said.
He and his colleagues are trying to encourage thoughtful, face-to-face conversations between physicians and patients about one of life's most intimate questions. But they successfully used the least intimate form of communication—social media—to tell the world that the ePrognosis website brought new information to bring to those conversations.
In fact, they called on a blogger—Paula Span, who writes “The New Old Age” for The New York Times—to help them seek reader input on how to create the website. Last year she posed a question to her readers on behalf of Lee and his colleagues: Should the ePrognosis site allow non-clinicians to access the prognostic information?
About 70 readers responded to her post, and they all shared an opinion. “They were fairly vehement about having access to such information,” Span wrote in a subsequent blog post. “Not one thought the site should be restricted to people with medical initials after their names.”
Writing on the GeriPal geriatrics and palliative care blog, Sei said that he and his colleagues worried that the average lay person is not trained to gauge which index is most appropriate for an individual situation, let alone how to interpret the statistical results. Despite being “deeply ambivalent,” they decided to allow non-clinicians to access the site—if they are willing to fib on the question “Are you a healthcare professional?,” which requires a “yes” before it will link into the questionnaire for a given index.
The deciding factor, he said, was that they trusted patients and caregivers to carry the prognosis conversation farther than physicians would.
“We believed that we'd have far greater potential for promoting a national—or international—conversation about this topic if we opened the site to patients and caregivers, encouraging them to discuss the information with their clinician,” he wrote.
He encourages oncologists to make a frank discussion of prognosis a central part of their interactions with their patients.
“My experience is that even though it's an uncomfortable topic, once you broach the subject and start talking about what the likely course is, that this is something that oftentimes they are already thinking about,” he said. “I have definitely had situations where it feels like the flood gates have opened and they are like, ‘Thank God someone is talking to me about this,’ and they are really grateful for the opportunity to discuss that with their physician.”
Listen to Dr. Steven Tucker discuss why and how he uses ePrognosis to put patients' cancer fears in perspective.
If you're not reading our iPad issues, download the free OT app (http://bit.ly/OT-iPadApp) and receive all 2012 issues to date.